SUMMARY Liquid-liquid phase separation (LLPS), i.e. the ability of molecules to condense into liquid-like assemblies, compartmentalizes cells extensively and impacts many fundamental biological processes. Whether LLPS is required for function in cells remains largely unclear. One challenge in answering this question arises from the difficulty in modulating the ability to form condensates without affecting the proteins' function, because assembly and function are often mediated by the same interactions. It is possible that smaller, discrete complexes are able to facilitate the function. We will address this question in enzymatically active condensates of the tumor suppressor speckle-type POZ protein (SPOP). SPOP recruits substrates to a ubiquitin ligase for ubiquitination. We have recently shown that SPOP and substrates undergo LLPS via weak, multivalent interactions, which result in their colocalization in active, membraneless organelles. Prostate cancer mutations blunt the ability of SPOP to phase separate with substrates, leading to their separate localization in cells, to increased substrate levels, and transformation of susceptible cells. We have experience in characterizing multivalent, disordered and phase-separating systems, and have built the necessary in vitro biophysical, biochemical and cell biological approaches and reagents to tackle the above question. In the proposed work, we will first modulate the material properties of condensates to test the requirement of fluidity for effective enzymatic activity. Second, we will test whether designed monovalent substrates, which bind at similar affinities as their multivalent counterparts, can be ubiquitinated effectively in the absence of phase separation. Third, we will make use of cancer mutations that modulate the formation of condensates and discrete complexes in opposite directions to test which of the two are the major players in SPOP function. Forth, we will address the critical question whether the weak interactions that typically mediate LLPS are able to compartmentalize cells specifically. We will use SPOP endometrial cancer mutations, which alter substrate specificity, to identify the strongest motifs responsible for the specificity alteration. The results will provide a conservative measure of specificity-mediating affinities in phase-separating systems. Our rigorous, multidisciplinary studies will significantly advance the knowledge of the structural determinants of specificity in weak SPOP/substrate interactions that drive phase separation, of the necessity of phase separation for SPOP-mediated substrate ubiquitination, and of the biophysical basis for the dysfunction of several SPOP cancer mutations that are distinct from the well-characterized prostate cancer mutations. The expected results will therefore provide conceptual insights into the role of phase separation in biological function. While we use rare cancer-associated mutations mainly as guides towards understanding of normal SPOP function, our work may ultimately help guide target validation for developing therapeutics against SPOP- related cancers.